I'm looking to perform a kernel density estimation on a set of 40 locations. At first I used LSCV as the bandwidth, however it seems to have oversmoothed the data, creating high densities in areas it shouldn't.
I've been looking around for information on which bandwidths are the best to use in which situations but am struggling. Does anyone know of a resource which explains this well?
Basically I'm hoping to find a bandwidth option which won't oversmooth the data as LSCV has and then be able justify my choice from the literature.
Just trying to revive this conversation, because if it has been answered well, I have not found it.
I have gone through the spatial stats walk through which describes using the Incremental Spatial Autocorrelation tool to find the distance with the most clustering, but the ranges it supplies are always so incredibly high, as to be almost completely useless. The ouput is always so generalized.
I also looked at just using the average nearest neighbor as the bandwidth, which is suggested in some literature, but I can't find much to provide for the justification of this.